In petroleum geology, exploration and production wells are often analyzed using image logs, because they provide a visual representation of the borehole surface and they are fundamental to retrieve information on bedding and rocks characteristics. In this paper, we present a novel approach for image log interpretation and extraction of the main features of the rock formation. This process led to the development of I2AM, a semiautomatic system that exploits image processing algorithms and artificial intelligence techniques to analyze and classify borehole images. I2AM analyzes log images using several image processing algorithms in order to extract numerical values for each characteristic and then performs a hierarchical clustering over the data obtained. Using three cluster evaluation indexes, possibly combined, I2AM can evaluate clustering results and, performing an automatic index-driven search, supplies a classification of the image logs. In this paper, we show I2AM application to the image logs from one well which were processed using the proposed method to identify different rock types and which were compared with those identified by the geologist. Main advantages of this approach are that interpretation time reduces from days to hours and subjectivity errors are avoided.

An AI tool for the petroleum industry based on image analysis and hierarchical clustering

FERRARETTI, Denis;GAMBERONI, Giacomo;LAMMA, Evelina;
2009

Abstract

In petroleum geology, exploration and production wells are often analyzed using image logs, because they provide a visual representation of the borehole surface and they are fundamental to retrieve information on bedding and rocks characteristics. In this paper, we present a novel approach for image log interpretation and extraction of the main features of the rock formation. This process led to the development of I2AM, a semiautomatic system that exploits image processing algorithms and artificial intelligence techniques to analyze and classify borehole images. I2AM analyzes log images using several image processing algorithms in order to extract numerical values for each characteristic and then performs a hierarchical clustering over the data obtained. Using three cluster evaluation indexes, possibly combined, I2AM can evaluate clustering results and, performing an automatic index-driven search, supplies a classification of the image logs. In this paper, we show I2AM application to the image logs from one well which were processed using the proposed method to identify different rock types and which were compared with those identified by the geologist. Main advantages of this approach are that interpretation time reduces from days to hours and subjectivity errors are avoided.
2009
3642043933;978-364204393-2
ARTIFICIAL INTELLIGENCE; DATA MINING; IMAGE PROCESSING; DENDROGRAM; HIERARCHICAL CLUSTERING; WELL; FMI IMAGES; CLUSTERING EVALUATION INDEXES
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in SFERA sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11392/1377562
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo

Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 8
  • ???jsp.display-item.citation.isi??? 5
social impact